Project Summary/Abstract Early-onset Alzheimer's disease (EOAD) is defined by an onset before age 65 and is characterized by the same neuropathology of late-onset Alzheimer's disease (LOAD). There is a common misconception that EOAD occurs primarily as an autosomal dominant disease, yet pathogenic variants in APP, PSEN1, and PSEN2 account for merely 5% of all EOAD cases. Only half or fewer EOAD patients carry the strong AD risk factor, APOE*E4. Nevertheless, family studies indicate that EOAD has a stronger heritable component than LOAD, suggesting a large portion of genetic risk remains unknown. Adding further complexity, unlike the typical episodic memory impairment of LOAD, EOAD often presents with ?atypical? clinical symptoms (i.e., executive, visuospatial, or language dysfunction) due to neurodegeneration of specific associated brain networks. Although LOAD and EOAD are defined by the same neuropathology, the fact that different underlying brain networks are affected suggests that EOAD results from a distinct underlying molecular etiology. The long-term goal of this work is to elucidate the genetic drivers of clinical heterogeneity in EOAD in order to develop predictive measures of individualized disease risk. In the present study, 900 EOAD, LOAD, and healthy age- matched controls will be studied through whole genome sequencing for novel genetic variation contributing to EOAD disease risk. Single-cell droplet-based RNA-sequencing will also be performed on a subset of patients to identify signatures of peripheral gene expression that distinguish EOAD patients. Finally, changes in gene expression will be related to patterns of brain atrophy to identify the group of genes contributing to selective neuroanatomical vulnerability in clinical subgroups of EOAD patients. The underlying hypothesis of this study is that there are different networks of genes linked by common biological pathways that drive selective vulnerability to each of the three brain networks that are vulnerable in different cases of EOAD. The specific aims of this project are: (1) Distinguish between EOAD genetic risk and LOAD genetic risk; (2) Identify gene expression differences between EOAD and LOAD; (3) Evaluate whether gene expression patterns predict EOAD atrophy patterns. Identifying underlying genetic risk contributing to EOAD clinical heterogeneity will inform our understanding of disease biology. In addition, predicting in advance which network is most vulnerable may enable us to identify the cognitive functions that should be monitored most closely during preclinical stages of disease in a patient-specific manner. The proposed cross-sectional study will lay the foundation for future work assessing the clinical value of using genomic, transcriptomic, and imaging profiles in combination to predict disease presentation and clinical progression in longitudinal cohorts of EOAD patients. This work will contribute to our biological understanding of variability in AD and may inform future efforts to develop personalized genomic medicine for EOAD prognostication and tracking during clinical intervention trials.